- Open Access
Modulation of walking speed by changing optic flow in persons with stroke
© Lamontagne et al; licensee BioMed Central Ltd. 2007
Received: 18 January 2007
Accepted: 26 June 2007
Published: 26 June 2007
Walking speed, which is often reduced after stroke, can be influenced by the perception of optic flow (OF) speed. The present study aims to: 1) compare the modulation of walking speed in response to OF speed changes between persons with stroke and healthy controls and 2) investigate whether virtual environments (VE) manipulating OF speed can be used to promote volitional changes in walking speed post stroke.
Twelve persons with stroke and 12 healthy individuals walked on a self-paced treadmill while viewing a virtual corridor in a helmet-mounted display. Two experiments were carried out on the same day. In experiment 1, the speed of an expanding OF was varied sinusoidally at 0.017 Hz (sine duration = 60 s), from 0 to 2 times the subject's comfortable walking speed, for a total duration of 5 minutes. In experiment 2, subjects were exposed to expanding OFs at discrete speeds that ranged from 0.25 to 2 times their comfortable speed. Each test trial was paired with a control trial performed at comfortable speed with matching OF. For each of the test trials, subjects were instructed to walk the distance within the same time as during the immediately preceding control trial. VEs were controlled by the CAREN-2 system (Motek). Instantaneous changes in gait speed (experiment 1) and the ratio of speed changes in the test trial over the control trial (experiment 2) were contrasted between the two groups of subjects.
When OF speed was changing continuously (experiment 1), an out-of-phase modulation was observed in the gait speed of healthy subjects, such that slower OFs induced faster walking speeds, and vice versa. Persons with stroke displayed weaker (p < 0.05, T-test) correlation coefficients between gait speed and OF speed, due to less pronounced changes and an altered phasing of gait speed modulation. When OF speed was manipulated discretely (experiment 2), a negative linear relationship was generally observed between the test-control ratio of gait speed and OF speed in healthy and stroke individuals. The slope of this relationship was similar between the stroke and healthy groups (p > 0.05, T-test).
Stroke affects the modulation of gait speed in response to changes in the perception of movement through different OF speeds. Nevertheless, the preservation of even a modest modulation enabled the persons with stroke to increase walking speed when presented with slower OFs. Manipulation of OF speed using virtual reality technology could be implemented in a gait rehabilitation intervention to promote faster walking speeds after stroke.
Regulation of walking involves the integration of visual, proprioceptive and vestibular information. Optic flow (OF) is a typical pattern of visual motion generated at the eye as the person moves through the environment . OF is a source of visual information that can be used to control heading direction [2–4] and speed [5–8] of walking, as it provides information about the direction and speed of self-motion with respect to the environment. Studies in healthy individuals have shown that changing OF speed modulates walking speed while walking on a self-paced treadmill [6–9]. During walking, visual (OF) and proprioceptive information about self-motion are normally congruent. When OF speed is manipulated in such way that it mismatches the proprioceptive information from the legs, walking velocity is adjusted to reduce the incongruity. The strategy adopted by healthy subjects is to reduce their walking speed with an increasing speed of OF, and to increase their speed at slower OFs [6, 7]. This adaptation has been shown to be altered by neurodegenerative diseases such as Parkinson's disease , in which the heavy reliance on visual  and kinesthetic information would produce exaggerated gait speed modulation responses.
Slow speed, a feature that characterizes locomotion after stroke , can impact on functional ability such as crossing a street within the time allotted by the pedestrian light. It is well accepted that muscle weakness, particularly on the paretic lower limb, is one of the main factors explaining slow gait speed after stroke [13, 14]. When motivated and instructed to walk at faster speeds, however, subjects with stroke have the capacity to at least double their walking speed , suggesting that factors other than muscle weakness may be contributing. It has also been shown that the discrimination of direction of global motion is altered by stroke, even when the primary cortical visual areas are preserved [16, 17]. Whether similar deficits in discriminating OF speed exist in this population is yet to be determined. The present study is based on the premise that an altered perception or integration of OF speed information could contribute to the difficulty of subjects with stroke in regulating walking speed. A residual ability to utilize OF to control the speed of walking, however, would provide the basis for gait intervention with virtual reality that manipulates OF parameters to enhance walking ability after stroke.
The specific aims of this study were: first, to compare the non-volitional modulation of walking speed in response to an OF of variable speed between persons with stroke and healthy controls and second, to investigate whether a virtual reality-based paradigm that manipulates OF speed could be used to promote volitional changes in walking speed in persons with stroke. We hypothesized that persons with stroke, although presenting with an altered modulation of walking speed in response to OFs of changing speed, would still present sufficient modulation such that a paradigm based on OF speed manipulation could be used to promote faster walking speeds.
Stroke (n = 12)
Time Stroke (mo)
0.60 (2 Canes)
(n = 12)
Set-up and Procedures
Subjects were first habituated to walk on the self-paced treadmill without the HMD, and then with the HMD while watching a virtual corridor scene expanding at a speed matching their comfortable walking speed. Comfortable walking speed was determined after a minimum of 4 to 5 minutes of habituation while walking with the HMD. Subjects listened to white and pink noises through earphones during testing, to avoid any speed feedback from auditory cues. White noise has equal energy at all frequencies, while pink noise has more power in the lower frequencies that the higher frequencies. The latter provides additional masking power for low frequency noises.
Two series of experiments were conducted on the same day, with experiment 2 following experiment 1. Because of fatigue or time constraint, one healthy control and 3 subjects with stroke did not complete experiment 2. In the first experiment, the instantaneous and non-volitional changes in walking speed in response to an OF of continually changing speed (Experiment 1) were examined. In the second, the voluntary or intentional changes in walking speed in response to OFs at discrete constant speeds (Experiment 2) were examined. In Experiment 1, the speed of an expanding OF was varied sinusoidally (0.017 Hz, 1 sine duration = 60 s), from 0 to 2 times the individual's comfortable walking speed, for a total duration of 5 minutes. The sine wave was preceded and ended by 30 seconds of OF matching comfortable speed. In Experiment 2, subjects were instructed to walk at a comfortable pace (control trials) on the self-paced treadmill through a 13 m-long virtual corridor comprising of 2 doorways located 10 m apart. Subjects started walking 1.5 m before the first door and ended at 1.5 m after the second door when a stop sign appeared. Every control trial was followed by a test speed trial, during which expanding OFs corresponding to 0.25, 0.5, 0.75, 1.00, 1.25, 1.50, 1.75 and 2 times the individual's comfortable speed were randomly presented. For each test speed trial, subjects were instructed to cover the corridor distance within the same time span as the previous control trial.
Distance travelled and walking speed were provided by the treadmill tachometer and recorded at 100 Hz using the Caren-2® system (Motek) that also manipulated the OF speeds. In addition, the treadmill speed signal was recorded at 360 Hz using the Vicon-512 A-D system. Speed signals were low-pass filtered at 0.02 Hz (second-order Butterworth, dual pass for zero lag).
For Experiment 1, the strength of the relationship and phase lag between the OF speed and walking speed signals were analyzed by means of cross-correlations. For Experiment 2, the mean walking speed over the 10 m separating the 2 doorways was calculated for every trial. Changes in walking speed for the speed trials were expressed as a ratio of the mean walking speed in the test trial over the control trial (gait speed ratio). The slopes between gait and OF speed ratios were obtained using linear regressions. All outcome variables, including cross-correlation coefficients, phase lags and slopes were compared between subjects with stroke and healthy controls using Student t-tests for independent samples. The relationship of these variables with the subjects' comfortable gait speeds was also quantified using Pearson correlation coefficients. Statistical analyses were carried out in Statistica 7.0 ® and the level of significance was set p < 0.05.
Discussion and Conclusions
The purpose of this study was to investigate the ability of persons with stroke to modulate their speed of walking in response to changing OF speeds. The main findings are that subjects with stroke present with alterations in their speed modulation response to changing OF speeds. However, when instructed to use OF speed information to scale their walking speed, they display patterns of modulation that are similar to those observed in age-matched healthy subjects.
Results of the present study demonstrate that walking speed is modulated out-of-phase with OF speed in healthy subjects, which is consistent with previous studies [6–8], despite the fact that expanding OFs were simulated by a meaningful and rich virtual environment in the present study. Previous research used gratings composed of dots, spots or diamonds oscillating to form contracting and expanding OFs [6–8]. The observed non-volitional changes in gait speed induced by changes in OF speed in our healthy age-matched controls, although consistent, were of small magnitude. According to Varraine and collaborators , this consistency combined with the fact that OF speed variations are not fully compensated by changes in gait speed, would argue in favor of a 'low level phenomenon' that is not under voluntary control. Present findings also revealed that gait speed responses were lagging behind the changes in OF speed. Such a lag may be explained by the latency of OF perception added to the latency for the mechanical transformation of the segmental kinematics.
Results from Experiment 2, in which subjects voluntarily attempted to walk a fixed distance within the same time span while exposed to OFs of different speeds, yielded larger walking speed changes than when exposed to the sinusoidally changing OF speeds in Experiment 1. This difference was observed despite the fact that the range of OF speed was smaller in Experiment 2 (0.25 to 2 times the individual's comfortable speed) than during Experiment 1 (0 to 2 times). The larger modulation observed in Experiment 2 may be attributed to the fact that information from constant OF speeds is easier to perceive and integrate than continually changing speeds, while a cognitive intentional process was also involved. The fact that the walking distance was shorter and that perception of time (duration required to walk the hallway) may have been a useful source of information to complete the task may also have contributed to the larger gait speed modulation responses observed in the second experiment. Gait speed adaptation did not fully compensate the mismatch with OF speed, even by our healthy aged-matched subjects, as indicated by slope values that did not approach -1.0. This partial compensation suggests the involvement of other sensorimotor transformations based on an internal representation and the central nervous system's integration of other sensory cues, such as leg proprioceptive information.
Present results also showed that subjects with stroke displayed abnormalities in their non-volitional or unintentional modulation of walking speed. Both animals [20–22] and human studies [23–26] suggest that optic flow perception involves not only the occipital, but also parietal and temporal cortical areas. Overall, as the complexity of the stimulus increases (first vs. second order stimuli), increasing neural networks involving more anterior brain regions are recruited [27–29]. Persons with a stroke in the temporal lobe or in adjacent regions of the frontal or parietal lobes manifest abnormally high thresholds for the discrimination of global motion direction . Whether they also present with higher discrimination thresholds for optic flow speed is still unclear. We have also shown that subjects with stroke present with a difficulty in integrating multiple sensory information, such as those induced during a rapid voluntary head turn while standing or walking [30, 31]. The altered modulation response observed in the subjects with stroke could thus be explained by an altered discrimination of OF speed and/or a reduced ability to integrate multiple sources of sensory information.
It is also noteworthy that the down-regulation response observed in the subjects with stroke differs from the up-regulation response observed in subjects with Parkinson's disease, which manifests as an abnormally large modulation in walking speed in response to changing OF speed . This finding in Parkinson's disease was interpreted as a higher reliance on visual kinesthesia, possibly due to an altered proprioceptive guidance of movement (proprioceptive kinesthesia). In the present study, none of our subjects with stroke showed abnormally large responses, which could suggest that they do not rely on visual kinesthesia as much as in Parkinson's disease. It is also possible that the characteristics of the OF substantially influences the speed modulation response. Hence, the use of an expanding OF in the present study may not be comparable to an OF that consists of both expansions and contractions in the study with Parkinson Disease subjects . In the early exploratory phase of this study, we had tested an OF that oscillated between expansion and contraction, simulating forward and backward self-motion. That type of stimulus caused a few subjects with stroke to behave differently in an attempt to walk backwards or completely stop walking with the contracting OFs (unpublished observations). For feasibility and safety reasons, as well as for the purpose of generalization of results to real life situations, it was decided in the present study that only expanding OFs simulating forward translation would be tested.
Results from Experiment 2 also reveal that subjects with stroke, once cognitively though intentionally involved in the task, produce speed modulation responses that are similar to that observed in the age-matched healthy subjects. The fact that the stroke group performed within the norms in the second experiment, but not during the first experiment, suggests that the two tasks involved different stimuli (changing vs. constant speed, reliance on time perception) and that incorporating a volitional component yields more accurate, yet not fully adapted, speed modulation responses. Most importantly, the adaptability of subjects with stroke in the second experiment provide a strong rationale to incorporate the manipulation of OF speed in a virtual-reality training paradigm that aims at promoting faster walking speeds. Present results indicate that subjects with stroke could instantaneously increase their walking speed by 1.44 times, which represents an increase close to 50%. To our knowledge, it is the first evidence that a virtual environment simulating OF could induce such large changes in gait speed in subjects with stroke. Before such an approach can be implemented in a virtual reality-based gait rehabilitation program, further work is needed to determine whether habituation occurs in the intentional speed modulation paradigm, as for the unintentional paradigm , and the carry-over effects to overground locomotion.
The authors would like to thank all participants, as well as Christian Beaudouin, Valeri Goussev, Luncinda Hughey, Eric Johnstone and Caroline Paquette for their technical assistance. The long corridor scene used in this experiment was adapted from the CAVE Library of VRCO, Inc. This study was funded by the Quebec Rehabilitation Research Network (REPAR) and the Canadian Institute of Health Research (CIHR – grant MOP-77548). Lamontagne is the recipient of a New Investigator Salary Award from CIHR.
- Gibson JJ: The visual perception of objective motion and subjective movement. 1954. Psychol Rev 1994,101(2):318-323. 10.1037/0033-295X.101.2.318View ArticlePubMedGoogle Scholar
- Warren WH Jr., Kay BA, Zosh WD, Duchon AP, Sahuc S: Optic flow is used to control human walking. Nat Neurosci 2001,4(2):213-216. 10.1038/84054View ArticlePubMedGoogle Scholar
- Warren WH: Perception of heading is a brain in the neck. Nat Neurosci 1998,1(8):647-649. 10.1038/3646View ArticlePubMedGoogle Scholar
- Harris JM, Bonas W: Optic flow and scene structure do not always contribute to the control of human walking. Vision Res 2002,42(13):1619-1626. 10.1016/S0042-6989(02)00066-4View ArticlePubMedGoogle Scholar
- Harris LR, Jenkin M, Zikovitz DC: Visual and non-visual cues in the perception of linear self-motion. Exp Brain Res 2000,135(1):12-21. 10.1007/s002210000504View ArticlePubMedGoogle Scholar
- Pailhous J, Ferrandez AM, Fluckiger M, Baumberger B: Unintentional modulations of human gait by optical flow. Behav Brain Res 1990,38(3):275-281. 10.1016/0166-4328(90)90181-DView ArticlePubMedGoogle Scholar
- Prokop T, Schubert M, Berger W: Visual influence on human locomotion. Modulation to changes in optic flow. Exp Brain Res 1997,114(1):63-70. 10.1007/PL00005624View ArticlePubMedGoogle Scholar
- Varraine E, Bonnard M, Pailhous J: Interaction between different sensory cues in the control of human gait. Exp Brain Res 2002,142(3):374-384. 10.1007/s00221-001-0934-3View ArticlePubMedGoogle Scholar
- Konzac J: Effects of the optic flow on the kinematics of the human gait: a comparaison of yong and older adults. J Mot Behav 1994, 26: 225-236.View ArticleGoogle Scholar
- Schubert M, Prokop T, Brocke F, Berger W: Visual kinesthesia and locomotion in Parkinson's disease. Mov Disord 2005,20(2):141-150. 10.1002/mds.20281View ArticlePubMedGoogle Scholar
- Barbeau H, Lamontagne A, Ladouceur M, Mercier I, Fung J: Optimizing locomotor function with body weight support training and functional electrical stimulation. In Progress in Motor Control III. Edited by: Latash M, Levin MF. Human Kinetics; 2003:237-251.Google Scholar
- von Schroeder HP, Coutts RD, Lyden PD, Billings E Jr., Nickel VL: Gait parameters following stroke: a practical assessment. J Rehabil Res Dev 1995,32(1):25-31.PubMedGoogle Scholar
- Nadeau S, Gravel D, Arsenault AB, Bourbonnais D: Plantarflexor weakness as a limiting factor of gait speed in stroke subjects and the compensating role of hip flexors. Clin Biomech (Bristol, Avon) 1999,14(2):125-135. 10.1016/S0268-0033(98)00062-XView ArticleGoogle Scholar
- Olney SJ, Griffin MP, McBride ID: Temporal, kinematic, and kinetic variables related to gait speed in subjects with hemiplegia: a regression approach. Phys Ther 1994,74(9):872-885.PubMedGoogle Scholar
- Lamontagne A, Fung J: Faster is better: implications for speed-intensive gait training after stroke. Stroke 2004,35(11):2543-2548. 10.1161/01.STR.0000144685.88760.d7View ArticlePubMedGoogle Scholar
- Vaina LM: Complex motion perception and its deficits. Curr Opin Neurobiol 1998,8(4):494-502. 10.1016/S0959-4388(98)80037-8View ArticlePubMedGoogle Scholar
- Vaina LM, Cowey A, Eskew RT Jr., LeMay M, Kemper T: Regional cerebral correlates of global motion perception: evidence from unilateral cerebral brain damage. Brain 2001,124(Pt 2):310-321. 10.1093/brain/124.2.310View ArticlePubMedGoogle Scholar
- Gauthier L, Dehaut F, Joanette Y: The Bells test for visual neglect: A quantitative and qualitative test for visual neglect. 1989.Google Scholar
- Halligan P, Wilson B, Cockburn J: A short screening test for visual neglect in stroke patients. Int Disabil Stud 1990,12(3):95-99.View ArticlePubMedGoogle Scholar
- Brosseau-Lachaine O, Faubert J, Casanova C: Functional sub-regions for optic flow processing in the posteromedial lateral suprasylvian cortex of the cat. Cereb Cortex 2001,11(10):989-1001. 10.1093/cercor/11.10.989View ArticlePubMedGoogle Scholar
- Page W, Duffy C: Heading representation in MST: sensory interactions and population encoding. J Neurophysiol 2003,89(4):1994-2013. 10.1152/jn.00493.2002View ArticlePubMedGoogle Scholar
- Steinmetz MA, Motter BC, Duffy CJ, Mountcastle VB: Functional properties of parietal visual neurons: radial organization of directionalities within the visual field. J Neurosci 1987,7(1):177-191.PubMedGoogle Scholar
- Greenlee MW: Human cortical areas underlying the perception of optic flow: brain imaging studies. Int Rev Neurobiol 2000, 44: 269-292.View ArticlePubMedGoogle Scholar
- Kleinschmidt A, Thilo KV, Buchel C, Gresty MA, Bronstein AM, Frackowiak RS: Neural correlates of visual-motion perception as object- or self-motion. Neuroimage 2002,16(4):873-882. 10.1006/nimg.2002.1181View ArticlePubMedGoogle Scholar
- Peuskens H, Sunaert S, Dupont P, Van Hecke P, Orban GA: Human brain regions involved in heading estimation. J Neurosci 2001,21(7):2451-2461.PubMedGoogle Scholar
- Ptito M, Kupers R, Faubert J, Gjedde A: Cortical representation of inward and outward radial motion in man. Neuroimage 2001,14(6):1409-1415. 10.1006/nimg.2001.0947View ArticlePubMedGoogle Scholar
- Mareschal I, Baker CL Jr.: Temporal and spatial response to second-order stimuli in cat area 18. J Neurophysiol 1998,80(6):2811-2823.PubMedGoogle Scholar
- Naidich TP, Daniels DL, Pech P, Haughton VM, Williams A, Pojunas K: Anterior commissure: anatomic-MR correlation and use as a landmark in three orthogonal planes. Radiology 1986,158(2):421-429.View ArticlePubMedGoogle Scholar
- Vaina LM, Solomon J, Chowdhury S, Sinha P, Belliveau JW: Functional neuroanatomy of biological motion perception in humans. Proc Natl Acad Sci U S A 2001,98(20):11656-11661. 10.1073/pnas.191374198PubMed CentralView ArticlePubMedGoogle Scholar
- Lamontagne A, De Serres S, Fung J, Paquet N: Stroke affects the coordination and stabilization of head, thorax and pelvis during voluntary horizontal head motions performed in walking. Clin Neurophys 2005,116(1):101-111. 10.1016/j.clinph.2004.07.027View ArticleGoogle Scholar
- Lamontagne A, Paquet N, Fung J: Postural adjustments to voluntary head motions during standing are modified following stroke. Clin Biomech (Bristol, Avon) 2003,18(9):832-842. 10.1016/S0268-0033(03)00141-4View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.